© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 32-TRANSCODER-HP
    7. 41092
    Prev
    Next
    INDEX
    Explanations

    Technical, scientific documents

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    绽æĶ¾
    -0.28
    eltas
    -0.28
     te
    -0.27
     deported
    -0.27
    bac
    -0.26
    pla
    -0.26
    oper
    -0.26
    .te
    -0.26
    ÑĤок
    -0.25
    èĵ¬
    -0.25
    POSITIVE LOGITS
     nghá»ĭ
    0.28
    anje
    0.26
    èĬĴ
    0.25
    urg
    0.25
    ctor
    0.24
    cast
    0.24
    mens
    0.24
    çĭĦ
    0.23
    irsch
    0.23
    midd
    0.23
    Activations Density 0.002%

    No Known Activations